A method and system for detecting the operating status of a display panel

Abstract

This invention discloses a method and system for detecting the operating status of a display panel, relating to the field of display panel operating status detection technology. The method includes preprocessing historical electrical signals and adjustment data collected by sensors, calculating the fluctuation amplitude of the electrical signal data and the response time of the adjustment data, constructing a decision matrix, obtaining weight coefficients for fluctuation amplitude and response time, defining an objective function for fusing the weight coefficients, outputting the optimal solution, optimizing based on the optimal solution, adjusting the sampling frequency for real-time sampling and denoising, and obtaining denoised data. This invention calculates the fluctuation amplitude of voltage and current, constructs a decision matrix based on the response time, uses the analytic hierarchy process (AHP) to assign relative importance to the state set, and further utilizes a genetic algorithm to minimize the fluctuation amplitude and response time, thereby effectively improving the accuracy and real-time performance of subsequent detection results. Furthermore, it significantly improves the accuracy and robustness of display panel operating status detection.

Description

Technical Field

[0001] This invention relates to the field of display panel operation status detection technology, and in particular to a method and system for detecting the operation status of a display panel. Background Technology

[0002] With the rapid development of modern display technology, display panels have been widely used in various electronic devices, including televisions, computer monitors, smartphones, and various industrial and medical devices. The operating status of display panels directly affects the performance of the devices and the user experience. In order to ensure the stable operation of display panels, especially in high-load or long-term use environments, real-time monitoring of their operating status is particularly important. With the continuous advancement of display technology, traditional manual inspection methods are no longer sufficient to meet the needs of efficient and accurate monitoring. Therefore, how to achieve accurate detection of the operating status of display panels through automation has become a hot topic in research and industry.

[0003] Traditional display panel operation status detection methods rely on simple filtering techniques when removing noise from data. These techniques cannot effectively eliminate high-frequency noise or extreme abnormal data, thus affecting the accuracy of the final status assessment. Furthermore, the accuracy and real-time performance of the detection results are insufficient to meet the requirements of high-standard application scenarios. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for detecting the operating status of a display panel, which solves the problem that traditional methods for detecting the operating status of display panels often rely on simple filtering techniques when removing noise from data, which cannot effectively remove high-frequency noise or extreme abnormal data, thus affecting the accuracy of the final status assessment. Furthermore, the accuracy and real-time performance of the detection results are difficult to meet the requirements of high-standard application scenarios.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for detecting the operating status of a display panel, comprising,

[0008] Historical electrical signals and regulation data are collected by sensors and preprocessed. The fluctuation amplitude of the electrical signal data and the response time of the regulation data are calculated separately. A decision matrix is ​​constructed to obtain the weight coefficients of fluctuation amplitude and response time. An objective function for fusing weight coefficients is defined, the optimal solution is output, and after optimization based on the optimal solution, the sampling frequency is adjusted for real-time sampling and denoising to obtain denoised data.

[0009] The electrical signals refer to voltage and current signals, and the adjustment data refers to the initial brightness and the adjusted target brightness of the display panel.

[0010] After decomposing the denoised data using variational mode decomposition, Hilbert transform is used to extract instantaneous energy and frequency features, and feature vectors are constructed.

[0011] Support vector machines are used as the classification model, and feature vectors are used as input to obtain classification results and confidence scores. After sorting the confidence scores, the panel states are obtained and stored.

[0012] As a preferred embodiment of the display panel operation status detection method of the present invention, the steps of calculating the fluctuation amplitude of electrical signal data and the response time of adjustment data, constructing a decision matrix, obtaining the weight coefficients of fluctuation amplitude and response time, defining the objective function of fusing the weight coefficients, and outputting the optimal solution include the following steps:

[0013] Based on standardized voltage and current data, calculate the mean of voltage and current, and then calculate the fluctuation range by taking the standard deviation.

[0014] The difference between the timestamps in the initial state and the target state is used as the response time, and then integrated with the fluctuation amplitude to obtain the state set at different time points;

[0015] After constructing the decision matrix based on the response time and fluctuation amplitude of all state sets, the relative importance between response time and fluctuation amplitude in the decision matrix is ​​obtained by using the proportional scaling method.

[0016] Using the analytic hierarchy process (AHP), after assigning values ​​to response time and volatility based on their relative importance, the average value of each row in the decision matrix is ​​calculated to obtain the weight coefficients for volatility and response time.

[0017] The fluctuation amplitude and response time within each state set are treated as individuals, and a population is generated for initialization.

[0018] Define an objective function to minimize the fluctuation range and response time;

[0019] The objective function value is used as the fitness value of each individual. The fitness values ​​of each individual are calculated and sorted in ascending order.

[0020] Set the selection threshold to ω, and compare the fitness value of each individual with the threshold ω to obtain the individuals to be retained;

[0021] Two rounds of competitive selection are performed on the retained individuals. The individual with the largest fitness value in the two rounds of competitive selection is selected, and after generating individual pairs, the operation is repeated in sequence until ω individual pairs are obtained. Then, the crossover probability of the individuals in the individual pairs is calculated.

[0022] Further, based on the individual's fitness value, the selection probability of each individual in the population is calculated;

[0023] Calculate the information entropy of each individual based on the selection probability of each individual;

[0024] The maximum information entropy among individuals is selected using the maximization operation, and the i-th individual S is... i The ratio of information entropy to maximum information entropy is used as an adjustment parameter. Then, the Sigmoid function is used to calculate the mutation probability of each individual;

[0025] A pseudo-random number generator is used to randomly generate crossover and mutation numbers for each individual.

[0026] The mutation probability of each individual is compared with the corresponding mutation random number to obtain the result of whether each individual has undergone mutation operation;

[0027] Next, the crossover probability of individuals in the selected individual pairs is compared with the corresponding crossover random number to obtain the result of whether the individual pairs should undergo crossover operation;

[0028] During the iteration process, the crossover probability and mutation probability are calculated based on the fitness value of each individual, and crossover and mutation operations are performed based on the crossover probability and mutation probability to generate new individuals. When the maximum number of iterations is reached, the minimized fluctuation range and response time are output.

[0029] As a preferred embodiment of the display panel operation status detection method of the present invention, the step of optimizing according to the optimal solution, adjusting the sampling frequency for real-time sampling and denoising, and obtaining denoised data includes the following steps:

[0030] Set the initial adjustment value and the scaling constant;

[0031] The target state is defined as the minimum fluctuation range and response time, and the difference between the target state and the corresponding data in the state set is used as the deviation value.

[0032] Using the proportional control formula, obtain the new adjustment value for fluctuation and the new adjustment value for response;

[0033] By using the exponential decay function in conjunction with the new adjustment value, the minimized volatility and response time are mapped to obtain the optimized volatility and response time;

[0034] Based on the optimized response time, the sampling frequency of the voltage and current sensors is adjusted using the Nyquist theorem;

[0035] Using the adjusted sampling frequency, real-time voltage and current data of the display panel are collected;

[0036] Based on the optimized fluctuation range, a high-pass or low-pass filter is selected to remove noise from the voltage and current data, resulting in denoised data.

[0037] As a preferred embodiment of the display panel operation status detection method of the present invention, the step of decomposing the denoised data using variational mode decomposition technology, extracting instantaneous energy and frequency features using Hilbert transform, and constructing feature vectors includes the following steps:

[0038] Based on the denoised data, a fast Fourier transform is used to convert it to obtain the frequency domain signal;

[0039] The frequency domain signal is transformed using the FFT function in Matlab, and the spectral peaks are then integrated to construct a spectral peak map.

[0040] After counting the number of peaks in the spectrum, the number of peaks is taken as the total number of modes;

[0041] Using the maximum and minimum operations, the highest and lowest frequencies in the spectrum are selected, and the difference is taken as the frequency bandwidth. The highest frequency is taken as the center frequency, and the ratio of the frequency bandwidth to the number of modes is taken as the bandwidth constraint value.

[0042] The frequency domain signal is decomposed using mode decomposition technology to obtain π mode functions;

[0043] Construct an objective function to minimize reconstruction error and frequency bandwidth;

[0044] During the iteration process, when the maximum number of iterations is reached, each modal function is output;

[0045] Each modal function refers to a signal with a different frequency response.

[0046] Using the sliding window technique, each modal function is divided, and the local mean within each window is calculated using the mean formula. The local mean is then arranged in chronological order, and the local means that are adjacent to each other in the arrangement are compared to obtain the local mean retention point.

[0047] Based on the retained points, a local mean curve is plotted, and the positive envelope is obtained as the actual value of the real part. After transformation, the expected value of the imaginary part is obtained.

[0048] The ratio between the expected value of the imaginary part and the actual value of the real part is used as the imaginary part coefficient.

[0049] By combining the expected value of the imaginary part, the actual value of the real part, and the coefficients of the imaginary part, an analytical signal is obtained;

[0050] Calculate the instantaneous amplitude based on the analytical signal;

[0051] The square of the absolute value of the instantaneous amplitude is taken as the instantaneous energy;

[0052] Further, based on the analytic signal, the phase of the analytic signal is calculated using the atan2 function;

[0053] Based on the phase of the analytic signal, calculate the time derivative of the phase as the instantaneous frequency;

[0054] By using feature splicing technology, all instantaneous energies and instantaneous frequencies are spliced ​​together to obtain feature vectors.

[0055] As a preferred embodiment of the display panel operation status detection method of the present invention, the following steps are taken: A support vector machine is used as a classification model, and the feature vector is used as input to obtain the classification result and confidence level. After sorting the confidence levels, the panel status index is obtained. Alternatively, a support vector machine is used as a classification model, and the feature vector is input into the classification model to output the classification result and the corresponding confidence level value. After sorting the confidence levels in ascending order, the classification result corresponding to the highest confidence level value is selected as the detection basis.

[0056] As a preferred embodiment of the display panel operation status detection method of the present invention, the storage refers to storing the classification results, feature vectors, minimized fluctuation amplitude and response time in CSV file format, adding corresponding timestamps to the CSV file and then storing the CSV file in a database.

[0057] As a preferred embodiment of the display panel operation status detection method of the present invention, the preprocessing of historical electrical signals and adjustment data collected by sensors includes the following steps:

[0058] Historical voltage, current, and regulation data at different points in time can be obtained from the display panel via the API interface;

[0059] Based on the adjustment data, timestamp the initial and target states of the adjustment data;

[0060] The adjustment data refers to the initial brightness and the adjusted target brightness of the display panel;

[0061] After removing outliers from all data using median filtering, standardization is then performed.

[0062] Secondly, the present invention provides a display panel operating status detection system, comprising,

[0063] The acquisition and processing module is used to acquire electrical signals and preprocess regulation data.

[0064] The calculation and output module is used to calculate the fluctuation amplitude of electrical signal data and the response time of regulation data, construct a decision matrix, obtain the weight coefficients of fluctuation amplitude and response time, define the objective function of fusing weight coefficients, and output the optimal solution.

[0065] The optimization and feedback module is used to optimize the optimal solution and readjust the sampling frequency for sampling and denoising to obtain denoised data;

[0066] The decomposition and extraction module is used to decompose the denoised data, extract instantaneous energy and frequency features, and construct feature vectors;

[0067] The classification and storage module uses a support vector machine as the classification model and takes the feature vector as input to obtain the classification result and confidence score. After sorting the confidence scores, the panel state is obtained and stored.

[0068] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the display panel operating state detection method as described in the first aspect of the present invention.

[0069] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the display panel operating state detection method as described in the first aspect of the present invention.

[0070] The beneficial effects of this invention are as follows: By calculating the fluctuation amplitude of voltage and current, constructing a decision matrix in conjunction with the response time, assigning relative importance to the state set using the analytic hierarchy process, and further utilizing a genetic algorithm to minimize and optimize the fluctuation amplitude and response time, the accuracy and real-time performance of subsequent detection results are effectively improved. Secondly, by combining variational mode decomposition and Hilbert transform techniques to extract instantaneous energy and frequency features, and using a support vector machine classification model to achieve accurate classification of the display panel state, this invention significantly improves the accuracy and robustness of display panel operating state detection. Attached Figure Description

[0071] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0072] Figure 1 This is a flowchart of the display panel operation status detection method in Example 1.

[0073] Figure 2 This is a structural diagram of the display panel operation status detection system in Example 1.

[0074] Figure 3 This is a flowchart of parameter optimization in Example 1.

[0075] Figure 4 This is a flowchart of feature extraction in Example 1. Detailed Implementation

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0077] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0078] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0079] Example 1, referring to Figures 1-4 This is the first embodiment of the present invention, which provides a method for detecting the operating status of a display panel, including the following steps:

[0080] S1. Historical electrical signals and regulation data are collected by sensors and preprocessed. The fluctuation amplitude of the electrical signal data and the response time of the regulation data are calculated respectively. A decision matrix is ​​constructed to obtain the weight coefficients of fluctuation amplitude and response time. The objective function of the fusion weight coefficients is defined, the optimal solution is output, and after optimization based on the optimal solution, the sampling frequency is adjusted for real-time sampling and denoising to obtain denoised data.

[0081] Specifically, the preprocessing of historical electrical signals and adjustment data collected by sensors includes the following steps:

[0082] Historical voltage, current, and regulation data at different points in time can be obtained from the display panel via the API interface;

[0083] Based on the adjustment data, timestamp the initial and target states of the adjustment data;

[0084] The adjustment data refers to the initial brightness and the adjusted target brightness of the display panel;

[0085] After removing outliers from all data using median filtering, standardization is then performed.

[0086] By acquiring historical voltage, current, and adjustment data from the display panel in real time, and combining this with timestamp and median filtering techniques, the accuracy and stability of data processing are significantly improved. The timestamp ensures the temporal order of each set of data, making the adjustment process more precise and traceable. Furthermore, median filtering effectively removes abnormal data, avoiding adjustment errors caused by noise or interference, thus ensuring the smoothness and reliability of the data. Secondly, standardized operations ensure the consistency of different datasets, enabling precise comparison and effective processing of each data point, further improving the accuracy of brightness adjustment.

[0087] Furthermore, the fluctuation amplitude of the electrical signal data and the response time of the regulation data are calculated separately, a decision matrix is ​​constructed, the weighting coefficients of fluctuation amplitude and response time are obtained, the objective function of fusing the weighting coefficients is defined, and the optimal solution is output, including the following steps:

[0088] Based on standardized voltage and current data, the mean values ​​of voltage and current are calculated using the mean formula, and then the standard deviations are taken. After obtaining the standard deviations of voltage and current, the fluctuation range is calculated.

[0089]

[0090] In the formula, Z(e) represents the fluctuation amplitude at time e, σ(C(e)) represents the standard deviation of the voltage data at time e, and σ(V(e)) represents the standard deviation of the current data at time e.

[0091] The difference between the timestamps in the initial state and the target state is used as the response time, and then integrated with the fluctuation amplitude to obtain the state set at different time points;

[0092] After constructing the decision matrix based on the response time and fluctuation amplitude of all state sets, the relative importance between response time and fluctuation amplitude in the decision matrix is ​​obtained by using the proportional scaling method.

[0093] Using the analytic hierarchy process (AHP), after assigning values ​​to response time and fluctuation amplitude according to relative importance, further standardization is performed on each column of the decision matrix.

[0094] The average value of each row in the decision matrix is ​​calculated using the mean formula, yielding the weighting coefficients for fluctuation amplitude and response time, respectively.

[0095] The fluctuation amplitude and response time within each state set are treated as individuals, and a population is generated for initialization.

[0096] Define an objective function to minimize the fluctuation range and response time:

[0097] B(S i )=δ·Z(S i )+β·D(S i )

[0098] In the formula, B(S) i ) represents the i-th individual S i The objective function value, where δ represents the weighting coefficient for fluctuation amplitude, Z(S i ) represents the i-th individual S i The fluctuation amplitude, β represents the weighting coefficient of response time, D(S) i ) represents the i-th individual S i Response time;

[0099] The objective function value is used as the fitness value of each individual. The fitness values ​​of each individual are calculated and sorted in ascending order.

[0100] Based on the accuracy requirements and relevant domain knowledge, a screening threshold of ω is set. The fitness value of each individual is compared with the threshold ω. If the fitness value of an individual is less than or equal to the threshold ω, the individual is excluded; otherwise, the individual is retained.

[0101] Based on the retained individuals, a two-round selection process is performed. The individual with the highest fitness value from the two rounds of selection is selected, and individual pairs are generated. This process is repeated until ω individual pairs are obtained. Then, the crossover probability of individuals in each individual pair is calculated.

[0102]

[0103] In the formula, G(S) i ) represents the i-th individual S i The crossover probability, B(S) i ) represents the i-th individual S i The fitness value, B(S) j ) represents the j-th individual S j fitness value;

[0104] Further, based on the individual's fitness value, calculate the selection probability of each individual in the population:

[0105]

[0106] In the formula, p(S) i ) represents the i-th individual S i The probability of selection, where N represents the total number of individuals;

[0107] Calculate the information entropy of each individual based on their selection probability:

[0108]

[0109] In the formula, H(S) i ) represents the i-th individual S i The information entropy, where log represents the logarithmic operation;

[0110] The maximum information entropy among individuals is selected using the maximization operation, and the i-th individual S is... i The ratio of information entropy to maximum information entropy is used as an adjustment parameter.

[0111] By combining the adjustment parameters and using the Sigmoid function, the mutation probability of each individual is calculated:

[0112]

[0113] In the formula, a(S) i ) represents the i-th individual S i The mutation probability, exp(·) represents the indicator function;

[0114] A pseudo-random number generator is used to randomly generate crossover and mutation numbers for each individual.

[0115] The mutation probability of each individual is compared with the corresponding random mutation number. If the mutation probability of an individual is greater than or equal to the corresponding random mutation number, the individual is mutated; otherwise, the individual is left as is.

[0116] Then compare the crossover probability of the individuals selected in the competition with the corresponding crossover random number;

[0117] When the crossover probabilities of corresponding individuals are all greater than or equal to the crossover random number, the individuals in the pair are crossovered to generate new individuals. When there is only one individual in the pair with a crossover probability greater than or equal to the crossover random number, the individual with a crossover probability greater than or equal to the crossover random number is retained as the next generation and participates in subsequent crossover operations. Individuals with a crossover probability less than the crossover random number are directly used as the next generation and do not participate in subsequent crossover operations. When the crossover probabilities of the individuals in the pair are all less than the corresponding crossover random number, they do not participate in crossover operations and are directly used as the next generation.

[0118] During the iteration process, the crossover probability and mutation probability are calculated based on the fitness value of each individual, and crossover and mutation operations are performed based on the crossover probability and mutation probability to generate new individuals. When the maximum number of iterations is reached, the minimized fluctuation range and response time are output.

[0119] By calculating fluctuation amplitude using standardized voltage and current data, this invention effectively eliminates data interference and improves calculation accuracy. The standard deviation calculation method not only accurately reflects signal fluctuations but also ensures robust data processing, laying a solid foundation for subsequent optimization. The analytic hierarchy process (AHP) is used to assign weights to fluctuation amplitude and response time, and the decision matrix is ​​optimized using these weights, improving the scientific and rational nature of the decision-making process. This ensures a balanced weight distribution across objectives and avoids excessive interference from subjective factors. Furthermore, the competition selection and fitness value screening mechanisms ensure the transmission of high-quality individuals during optimization, filtering out unfit individuals, accelerating algorithm convergence, and improving the overall population quality. The competition selection mechanism, by granting priority crossover opportunities to individuals with higher fitness, effectively avoids the inefficiency that might result from completely random selection. Secondly, the crossover probability... The calculation is based on individual fitness values, ensuring that individuals with high fitness are more likely to participate in crossover operations, further promoting the transmission of superior genes and optimizing the effect of genetic operations. Furthermore, the introduction of information entropy further enhances population diversity, preventing the algorithm from getting trapped in local optima during the search process. Through the calculation of information entropy, the mutation probability can be dynamically adjusted, ensuring the availability of individuals with excellent fitness while introducing new gene mutations in a timely manner, avoiding premature convergence. This allows the invention to flexibly adjust the search strategy according to the actual situation at different stages, enhancing global search capabilities. By dynamically adjusting the operation probabilities of crossover and mutation, a balance between exploration and development during the search process is ensured. Finally, the use of the sigmoid function regulates the mutation probability, establishing an effective link between the intensity of mutation and individual fitness and population diversity, improving the algorithm's adaptability and robustness.

[0120] Furthermore, after optimization based on the optimal solution, the sampling frequency is adjusted for real-time sampling and denoising to obtain denoised data, including the following steps:

[0121] Initial adjustment values ​​and proportional constants are set based on experience and actual needs;

[0122] The target state is defined as the minimum fluctuation range and response time, and the difference between the target state and the corresponding data in the state set is used as the deviation value.

[0123] Using the proportional control formula, the difference between the initial adjustment value and the product of the proportional constant and the corresponding deviation value is used as the new adjustment value for the fluctuation amplitude and the response time, respectively, thus obtaining the new adjustment value for fluctuation and the new adjustment value for response.

[0124] By using the exponential decay function in conjunction with the new adjustment value, the minimized volatility and response time are mapped to obtain the optimized volatility and response time:

[0125]

[0126] In the formula, F represents the optimized fluctuation range and response time, e represents the base of the natural logarithm, θ1 represents the new adjusted value of the fluctuation range, and θ2 represents the new adjusted value of the response time;

[0127] Based on the optimized response time, the sampling frequency of the voltage and current sensors is adjusted using the Nyquist theorem;

[0128] Using the adjusted sampling frequency, real-time voltage and current data of the display panel are collected;

[0129] Based on the optimized fluctuation range, a high-pass or low-pass filter is selected to remove noise from the voltage and current data, resulting in denoised data.

[0130] By applying deviation calculation and proportional control formulas, the new adjustment values ​​for fluctuation amplitude and response time are adjusted in real time, allowing the invention to remain near the target state, thereby improving optimization accuracy. Furthermore, the introduction of an exponential decay function smooths the optimization process, avoiding instability caused by over-adjustment and ensuring a reasonable range for fluctuation amplitude and response time. Regarding sampling frequency adjustment, the invention dynamically adjusts the sampling frequency based on the optimized response time according to the Nyquist theorem, ensuring data acquisition accuracy and avoiding signal distortion. Secondly, by selecting appropriate high-pass or low-pass filters to remove noise from the data, the denoising effect is significantly improved, making the final data clearer and more reliable. This invention achieves precise optimization and stable operation through a combination of feedback mechanisms, dynamic adjustment, and flexible denoising.

[0131] S2. After decomposing the denoised data using variational mode decomposition, Hilbert transform is used to extract instantaneous energy and frequency features, and feature vectors are constructed.

[0132] Specifically, after decomposing the denoised data using variational mode decomposition, the Hilbert transform is used to extract instantaneous energy and frequency features, and feature vectors are constructed, including the following steps:

[0133] Based on the denoised data, a fast Fourier transform is used to convert it to obtain the frequency domain signal;

[0134] The frequency domain signal is transformed using the FFT function in Matlab to obtain the spectral peak value;

[0135] Based on the spectral peak values, a spectral peak map is constructed.

[0136] Using statistical methods, the number of peaks in the spectrum is counted, and the number of peaks is taken as the total number of modes;

[0137] Using the maximum and minimum operations, the highest and lowest frequencies in the spectrum are selected, and the difference is taken as the frequency bandwidth, with the highest frequency being taken as the center frequency.

[0138] Furthermore, based on the frequency bandwidth, the ratio of frequency bandwidth to the number of modes is used as the bandwidth constraint value;

[0139] The frequency domain signal is decomposed using mode decomposition technology to obtain π mode functions;

[0140] Construct an objective function to minimize reconstruction error and frequency bandwidth:

[0141]

[0142] In the formula, u k (t) represents the k-th mode function at time t, w k Let K represent the center frequency associated with the k-th mode function, and K represent the total number of modes. Let O(t) denote the derivative with respect to time t, and let O(t) represent the frequency domain signal at time t. 2 Represents the square of the absolute value. Indicates the bandwidth constraint value, l k (t) represents the frequency bandwidth associated with the k-th mode function at time t;

[0143] During the iteration process, when the maximum number of iterations is reached, each modal function is output;

[0144] Each mode function refers to the signal with a different frequency response.

[0145] By using the sliding window technique, each modal function is divided, and the local mean within each window is calculated using the mean formula.

[0146] Arrange all local means in chronological order, and compare the local means that are adjacent to each other in the arrangement. If the selected local mean is greater than the adjacent local means, then retain the local mean; otherwise, discard the local mean.

[0147] Further, based on the arrangement results, the local mean of the first and last points in the arrangement results is compared with the local mean of their respective neighbors. If the local mean of the first and last points is greater than the local mean of their respective neighbors, the first and last points are retained; otherwise, the first and last points are removed.

[0148] Plot the local mean curve based on the retained points to obtain the positive envelope as the actual value of the real part;

[0149] The Hilbert transform of the real part is performed using the hilbert(.) function in MATLAB to obtain the expected value of the imaginary part;

[0150] The ratio between the expected value of the imaginary part and the actual value of the real part is used as the imaginary part coefficient.

[0151] By combining the expected value of the imaginary part, the actual value of the real part, and the coefficients of the imaginary part, the analytic signal is obtained:

[0152] A k (t)=M k (t)+mE{M k (t)}

[0153] In the formula, A k (t) represents the analytic signal of the k-th mode function at time t, M k (t) represents the actual value of the real part of the k-th mode function at time t, E{M k (t)} represents the expected value of the imaginary part of the k-th mode function at time t, and m represents the imaginary part coefficient;

[0154] Calculate the instantaneous amplitude based on the analytical signal:

[0155]

[0156] In the formula, o k (t) represents the instantaneous amplitude of the k-th mode function at time t;

[0157] The square of the absolute value of the instantaneous amplitude is taken as the instantaneous energy;

[0158] Further, based on the analytic signal, the phase of the analytic signal is calculated using the atan2 function:

[0159]

[0160] In the formula, arg(A) k (t) represents the analytic signal phase of the k-th mode function at time t. Indicates the real part, Indicates the imaginary part;

[0161] Based on the phase of the analytic signal, the time derivative of the phase is calculated as the instantaneous frequency:

[0162]

[0163] In the formula, y k (t) represents the instantaneous frequency of the k-th mode function at time t. Represents the time derivative;

[0164] By using feature splicing technology, all instantaneous energies and instantaneous frequencies are spliced ​​together to obtain feature vectors.

[0165] Variational mode decomposition (VMD) technology decomposes complex signals into multiple modal components, removing high-frequency noise while retaining important frequency characteristics, significantly improving signal quality. Hilbert transform is used to extract the instantaneous amplitude and frequency of the signal. Instantaneous frequency reveals the signal's frequency variation characteristics, while instantaneous amplitude provides information on the signal strength change over time. This allows for precise capture of the signal's dynamic characteristics during feature extraction. The sliding window technique is used to calculate local means, effectively removing insignificant local features. By comparing the adjacent relationships of local means, more representative features are retained, improving feature extraction accuracy. Furthermore, the introduction of spectral analysis and bandwidth constraints further optimizes the signal's frequency range definition, ensuring efficiency and accuracy during reconstruction. Finally, instantaneous energy and instantaneous frequency are concatenated into a feature vector, facilitating subsequent analysis.

[0166] S3. Use support vector machine as the classification model and feature vector as input to obtain classification results and confidence scores. After sorting the confidence scores, obtain the panel state and store it.

[0167] Specifically, a support vector machine (SVM) is used as the classification model, and the feature vector is used as input to obtain the classification result and confidence score. After sorting the confidence scores, the panel state index is obtained. Using a support vector machine as the classification model, the feature vector is input into the classification model, and the classification result and corresponding confidence score value are output. After sorting the confidence scores in ascending order, the classification result corresponding to the highest confidence score value is selected as the detection basis.

[0168] The process of building and training the classification model is as follows:

[0169] The width parameter and Lagrange multipliers are set based on personal experience and knowledge in the relevant field;

[0170] The RBF kernel function is used in conjunction with a width parameter to map the feature vectors, thus obtaining the kernel function value;

[0171] Use support vector machines as the classification model;

[0172] Combining the kernel function value and the Lagrange multiplier, we define the objective function and maximize the objective function:

[0173]

[0174] In the formula, X(a) represents the objective function value associated with the Lagrange multiplier α, R represents the total number of samples, and α q Let α represent the Lagrange multiplier of the q-th sample. z Let x represent the Lagrange multiplier of the z-th sample. q Let x represent the feature vector of the q-th sample. zLet J(s) represent the feature vector of the z-th sample. q ,s z ) represents the kernel function value, s q Let s represent the q-th sample. z This represents the z-th sample;

[0175] The model parameters are updated using a quadratic programming solver. During the iteration process, the iteration stops when the maximum number of iterations is reached, and the final classification model is output.

[0176] The feature vector is output to the classification model to obtain the classification result and confidence value.

[0177] By inputting feature vectors, the Support Vector Machine (SVM) can output classification results and confidence values, improving the accuracy and reliability of classification. Furthermore, by ranking the confidence scores and selecting the classification result corresponding to the highest confidence score, the possibility of misclassification is further reduced. Secondly, the RBF kernel function maps the data to a high-dimensional space, solving the nonlinearity problem and enabling SVM to handle more complex real-world data. The introduction of Lagrange multipliers optimizes the model's classification boundary and improves the classifier's generalization ability by maximizing the objective function, avoiding overfitting. Thirdly, the quadratic programming solver effectively optimizes model parameters, ensuring accurate selection of support vectors and thus improving the performance of the classification model. Finally, by combining confidence ranking with classification results, this invention effectively improves the efficiency and accuracy of display panel operation status detection.

[0178] Furthermore, storage refers to storing the classification results, feature vectors, minimized fluctuation range, and response time in CSV file format, adding corresponding timestamps to the CSV file, and then storing the CSV file using a database.

[0179] By storing classification results, feature vectors, fluctuation amplitudes, and response times in CSV file format, and adding timestamps to each file before storing them in a database, efficient data management and access are achieved. This invention enhances data traceability, the timestamps ensure the timeliness of each record and facilitate historical data tracing, and the database storage enables rapid querying and analysis of large-scale data.

[0180] This embodiment also provides a display panel operation status detection system, including:

[0181] The acquisition and processing module is used to acquire electrical signals and preprocess regulation data.

[0182] The calculation and output module is used to calculate the fluctuation amplitude of electrical signal data and the response time of regulation data, construct a decision matrix, obtain the weight coefficients of fluctuation amplitude and response time, define the objective function of fusing weight coefficients, and output the optimal solution.

[0183] The optimization and feedback module is used to optimize the optimal solution and readjust the sampling frequency for sampling and denoising to obtain denoised data;

[0184] The decomposition and extraction module is used to decompose the denoised data, extract instantaneous energy and frequency features, and construct feature vectors;

[0185] The classification and storage module uses a support vector machine as the classification model and takes the feature vector as input to obtain the classification result and confidence score. After sorting the confidence scores, the panel state is obtained and stored.

[0186] This embodiment also provides a computer device applicable to the display panel operating status detection method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the display panel operating status detection method proposed in the above embodiment.

[0187] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0188] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the display panel operation status detection method as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0189] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for detecting the operating status of a display panel, characterized in that: include, Historical electrical signals and regulation data are collected by sensors and preprocessed. The fluctuation amplitude of the electrical signal data and the response time of the regulation data are calculated separately. A decision matrix is ​​constructed to obtain the weight coefficients of fluctuation amplitude and response time. An objective function for fusing weight coefficients is defined, the optimal solution is output, and after optimization based on the optimal solution, the sampling frequency is adjusted for real-time sampling and denoising to obtain denoised data. The electrical signals refer to voltage and current signals, and the adjustment data refers to the initial brightness and the adjusted target brightness of the display panel. After decomposing the denoised data using variational mode decomposition, Hilbert transform is used to extract instantaneous energy and frequency features, and feature vectors are constructed. Support vector machines are used as the classification model, and feature vectors are taken as input. The output is the classification result and the corresponding confidence value. After sorting the confidence values ​​in ascending order, the classification result with the highest confidence value is selected as the detection basis and stored.

2. The display panel operating status detection method as described in claim 1, characterized in that: The steps involved in calculating the fluctuation amplitude of the electrical signal data and the response time of the regulation data, constructing a decision matrix, obtaining the weighting coefficients for fluctuation amplitude and response time, defining the objective function for fusing the weighting coefficients, and outputting the optimal solution include: Based on standardized voltage and current data, calculate the mean of voltage and current, and then calculate the fluctuation range by taking the standard deviation. The difference between the timestamps in the initial state and the target state is used as the response time, and then integrated with the fluctuation amplitude to obtain the state set at different time points; After constructing the decision matrix based on the response time and fluctuation amplitude of all state sets, the relative importance between response time and fluctuation amplitude in the decision matrix is ​​obtained by using the proportional scaling method. Using the analytic hierarchy process (AHP), after assigning values ​​to response time and volatility based on their relative importance, the average value of each row in the decision matrix is ​​calculated to obtain the weight coefficients for volatility and response time. The fluctuation amplitude and response time within each state set are treated as individuals, and a population is generated for initialization. Define an objective function to minimize the fluctuation range and response time; The objective function value is used as the fitness value of each individual. The fitness values ​​of each individual are calculated and sorted in ascending order. Set the selection threshold to ω, and compare the fitness value of each individual with the threshold ω to obtain the individuals to be retained; Two rounds of competitive selection are performed on the retained individuals. The individual with the largest fitness value in the two rounds of competitive selection is selected, and after generating individual pairs, the operation is repeated in sequence until ω individual pairs are obtained. Then, the crossover probability of the individuals in the individual pairs is calculated. Further, based on the individual's fitness value, the selection probability of each individual in the population is calculated; Calculate the information entropy of each individual based on the selection probability of each individual; The maximum information entropy among individuals is selected using the maximization operation, and the i-th individual S is... i The ratio of information entropy to maximum information entropy is used as an adjustment parameter. Then, the Sigmoid function is used to calculate the mutation probability of each individual; A pseudo-random number generator is used to randomly generate crossover and mutation numbers for each individual. The mutation probability of each individual is compared with the corresponding mutation random number to obtain the result of whether each individual has undergone mutation operation; Next, the crossover probability of individuals in the selected individual pairs is compared with the corresponding crossover random number to obtain the result of whether the individual pairs should undergo crossover operation; During the iteration process, the crossover probability and mutation probability are calculated based on the fitness value of each individual, and crossover and mutation operations are performed based on the crossover probability and mutation probability to generate new individuals. When the maximum number of iterations is reached, the minimized fluctuation range and response time are output.

3. The display panel operating status detection method as described in claim 2, characterized in that: After optimization based on the optimal solution, the sampling frequency is adjusted for real-time sampling and denoising to obtain denoised data, which includes the following steps: Set the initial adjustment value and the scaling constant; The target state is defined as the minimum fluctuation range and response time, and the difference between the target state and the corresponding data in the state set is used as the deviation value. Using the proportional control formula, obtain the new adjustment value for fluctuation and the new adjustment value for response; By using the exponential decay function in conjunction with the new adjustment value, the minimized fluctuation amplitude and response time are mapped to obtain the optimized fluctuation amplitude and response time; Based on the optimized response time, the sampling frequency of the voltage and current sensors is adjusted using the Nyquist theorem; Using the adjusted sampling frequency, real-time voltage and current data of the display panel are collected; Based on the optimized fluctuation range, a high-pass or low-pass filter is selected to remove noise from the voltage and current data, resulting in denoised data.

4. The display panel operation status detection method as described in claim 3, characterized in that: The process of decomposing the denoised data using variational mode decomposition (VMD) and then extracting instantaneous energy and frequency features using Hilbert transform to construct feature vectors includes the following steps: Based on the denoised data, a fast Fourier transform is used to convert it to obtain the frequency domain signal; The frequency domain signal is transformed using the FFT function in Matlab, and the spectral peaks are then integrated to construct a spectral peak map. After counting the number of peaks in the spectrum, the number of peaks is taken as the total number of modes; Using the maximum and minimum operations, the highest and lowest frequencies in the spectrum are selected, and the difference is taken as the frequency bandwidth. The highest frequency is taken as the center frequency, and the ratio of the frequency bandwidth to the number of modes is taken as the bandwidth constraint value. The frequency domain signal is decomposed using mode decomposition technology to obtain π mode functions; Construct an objective function to minimize reconstruction error and frequency bandwidth; During the iteration process, when the maximum number of iterations is reached, each modal function is output; Each modal function refers to a signal with a different frequency response. Using the sliding window technique, each modal function is divided, and the local mean within each window is calculated using the mean formula. The local mean is then arranged in chronological order, and the local means that are adjacent to each other in the arrangement are compared to obtain the local mean retention point. Based on the retained points, a local mean curve is plotted, and the positive envelope is obtained as the actual value of the real part. After transformation, the expected value of the imaginary part is obtained. The ratio between the expected value of the imaginary part and the actual value of the real part is used as the imaginary part coefficient. By combining the expected value of the imaginary part, the actual value of the real part, and the coefficients of the imaginary part, an analytical signal is obtained; Calculate the instantaneous amplitude based on the analyzed signal; The square of the absolute value of the instantaneous amplitude is taken as the instantaneous energy; Further, based on the analytic signal, the phase of the analytic signal is calculated using the atan2 function; Based on the phase of the analytic signal, calculate the time derivative of the phase as the instantaneous frequency; By using feature splicing technology, all instantaneous energies and instantaneous frequencies are spliced ​​together to obtain feature vectors.

5. The display panel operation status detection method as described in claim 4, characterized in that: The storage process involves storing the classification results, feature vectors, minimized fluctuation range, and response time in CSV file format, adding corresponding timestamps to the CSV file, and then storing the CSV file using a database.

6. The display panel operation status detection method as described in claim 5, characterized in that: The preprocessing of historical electrical signals and adjustment data collected by sensors includes the following steps: Historical voltage, current, and regulation data at different points in time can be obtained from the display panel via the API interface; Based on the adjustment data, timestamp the initial and target states of the adjustment data; After removing outliers from all data using median filtering, standardization is then performed.

7. A display panel operation status detection system, based on the display panel operation status detection method according to any one of claims 1 to 6, characterized in that: include, The acquisition and processing module is used to acquire electrical signals and preprocess regulation data. The calculation and output module is used to calculate the fluctuation amplitude of electrical signal data and the response time of regulation data, construct a decision matrix, obtain the weight coefficients of fluctuation amplitude and response time, define the objective function of fusing weight coefficients, and output the optimal solution. The optimization and feedback module is used to optimize the optimal solution and readjust the sampling frequency for sampling and denoising to obtain denoised data; The decomposition and extraction module is used to decompose the denoised data, extract instantaneous energy and frequency features, and construct feature vectors; The classification and storage module uses a support vector machine as the classification model and takes the feature vector as input to obtain the classification result and confidence score. After sorting the confidence scores, the panel state is obtained and stored.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the display panel operation status detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the display panel operation status detection method according to any one of claims 1 to 6.

Images